Laird Harrison

September 11, 2017

BARCELONA — The use of optical coherence tomography (OCT) to screen everyone with diabetes for retinopathy appears to be a relatively inexpensive way to preserve the quality of life of these patients, results from a pilot study demonstrate.

In fact, when used in combination with fundus photography, the sensitivity and specificity to detect the disease were both 100%.

"I was surprised," investigator Ian Wong, MBBS, from the University of Hong Kong, said here at the 17th EURETINA Congress.

Protocols vary widely from one country and healthcare system to another, but most rely on fundus photography, according to session moderator Jean-François Korobelnik, MD, from the University of Bordeaux in France. "It's rare to screen patients with OCT, but it makes a lot of sense," he said.

Although it would be ideal to screen everyone with diabetes, few countries have come close to that goal, even with fundus photography. In the United Kingdom, which is among the leaders in this area, about 80% of people with diabetes are screened, Dr Korobelnik reported.

It's rare to screen patients with OCT, but it makes a lot of sense.

Hong Kong has adopted the approach used in the United Kingdom, where screening is free for people being seen in general diabetes outpatient clinics. Optometrists use fundus photos of the macula and the area around the optic disc to look for diabetic retinopathy and diabetic macular edema, and refer patients to ophthalmologists when they suspect maculopathy.

However, because these photos provide only a two-dimensional view of the macula, it is difficult to visualize any edema. Often, optometrists rely on macular exudates and hemorrhages as surrogate markers of the disease.

But this system is not completely accurate, so patients might not receive the treatment they require and ophthalmologists might waste time on examinations for people who do not have macular edema, Dr Wong explained.

OCT provides a view of the macula in a third dimension. It is noninvasive, fast, repeatable, and quantitative, he pointed out. Still, the machines are expensive and the resources of operators and image readers are required for the additional imaging.

Dr Wong and his colleague wanted to know whether the cost of adding OCT to the screening process would be offset by improvements in quality of life for patients who might otherwise not be diagnosed as early.

For their study, the team added OCT to the routine screening of 2277 patients at the University of Hong Kong from February 2014 to January 2016.

They used their data to estimate the hypothetical costs of four different screening strategies.

Strategy A Fundus photography with retinal hemorrhage and exudates used as surrogate markers (existing protocol)
Strategy B Strategy A without the use of retinal hemorrhage as a surrogate marker
Strategy C Strategy A plus an OCT scan only if diabetic macular edema is suspected
Strategy D Strategy A plus an OCT scan in all cases


The investigators interviewed program managers and local experts from different hospitals to estimate the cost of the screening program, taking into account the cost of equipment and staffing, examinations done by ophthalmologists, adjunct investigations, and the treatment of diabetic macular edema for 1 year for patients correctly referred.

In their model, strategy D — fundus photography plus OCT for all patients — was the most expensive, but it caught all cases of diabetic macular edema, which resulted in a significantly greater quality-adjusted life-year (QALY).

Table 1. Accuracy of the Screening Strategies

Variable Strategy A Strategy B Strategy C Strategy D
Sensitivity, % 40.95 22.86 32.38 100.00
Specificity, % 86.60 95.63 100.00 100.00
Positive predictive value, % 12.87 20.10 100.00 100.00
Negative predictive value, % 96.81 96.25 96.30 100.00


Dr Wong said he was struck by the fact that only 12.87% of those who would have been referred to ophthalmologists under strategy A, the current standard, actually had diabetic macular edema. "Seven in eight were wrongly referred," he pointed out. "That results in wastage."

Over a longer period of time, the cost per QALY would likely be even lower with strategy D, he told Medscape Medical News.

Table 2. Cost and Benefit of the Screening Strategies

Benefit Strategy A Strategy B Strategy C Strategy D
QALY gained 19.4983 10.8828 15.4173 47.6123
Cost per QALY (US$) 7447.50 8428.70 5992.30 4113.50


The research team is planning to extend their study for 5 years.

After the presentation, a member of the audience asked why strategies B and C were so inefficient.

In those scenarios, "the way we selected patients to offer them OCT has to be further refined," Dr Wong acknowledged.

Any measure that can catch diabetic macular edema early would provide a clear benefit for patients, Dr Korobelnik told Medscape Medical News.

One challenge would be finding the resources to perform so many OCT scans. "If you can screen thousands of patients, can you look at all the images with your own eyes?" he wondered.

But that task could one day fall to artificial intelligence systems under development, he pointed out.

The study was funded by the Hong Kong Food and Health Bureau and the Hong Kong Medical Research Fund. Dr Wong has disclosed no relevant financial relationships. Dr Korobelnik is a consultant for Alcon, Allergan, Bayer, Boehringer-Ingelheim, Kanghong, Krys, Novartis, Roche, Thea, and Zeiss.

European Society of Retina Specialists 17th EURETINA Congress. Presented September 8, 2017.

Follow Medscape on Twitter @Medscape and Laird Harrison @LairdH


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.